File Example_A.txt

This is a typical PERMAP data file using similarities.
Any comments, such as this one, that do not start with a
reserved word are ignored.  The reserved words in this file
are typed in all caps.

Permap reads only text files, so if this is an Excel file save it as a MS-DOS text file.
Doing so may cause some quotes and strange characters to be added by 
the Excel save routine.  They don't hurt the use of the file.   Just ignore them.


TITLE= Ekman's Classic Color Similarity Data
NOBJECTS= 14

SIMILARITYLIST
a 	1													
b 	0.86	1												
c 	0.42	0.5	1											
d 	0.42	0.44	0.81	1										
e 	0.18	0.22	0.47	0.54	1									
f 	0.06	0.09	0.17	0.25	0.61	1								
g 	0.07	0.07	0.1	0.1	0.31	0.62	1							
h 	0.04	0.07	0.08	0.09	0.26	0.45	0.73	1						
i 	0.02	0.02	0.02	0.02	0.07	0.14	0.22	0.33	1					
j 	0.07	0.04	0.01	0.01	0.02	0.08	0.14	0.19	0.58	1				
k 	0.09	0.07	0.02	0.00	0.02	0.02	0.05	0.04	0.37	0.74	1			
l 	0.12	0.11	0.01	0.01	0.01	0.02	0.02	0.03	0.27	0.5	0.76	1		
m 	0.13	0.13	0.05	0.02	0.02	0.02	0.02	0.02	0.2	0.41	0.62	0.85	1	
n 	0.16	0.14	0.03	0.04	0	0.01	0	0.02	0.23	0.28	0.55	0.68	0.76	1

Ekman's data (1954) are for human perceptions 													
of the similarities of 14 different pure colors. The 
colors differ in wavelength but not in brightness 
or saturation. Ekman asked for judgments on the 
91 color pairs from 31 subjects using a 0 (no similarity) 
to 4 (identical) scale. The results from all 31 subjects 
were averaged and divided by 4 to get the similarity 
matrix.

This data set is often analyzed using an ordinal 
(nonmetric) approach, but ratio (metric) or interval (metric) 
approaches work just as well. Sometimes nonmetric 
analyses make good theoretical sense and 
sometimes they are the sign of a lazy experimentalist 
or timid analyst. 

This data set is interesting historically and provides a  
good example of a typical MDS application, even though Ekman 
did not use MDS in its analysis. However, 
there is some question about the validity of 
the conclusions that were drawn based on the data. 
Please see the PERMAP operations manual for more 
information on conflicts between large and small dissimilarity 
values, the occurrence of circular patterns, 
and testing by relaxation into a higher dimension. 




